Feedback-related midfrontal theta power.

(a) Left: Time-wise permutation testing reveals one feedback-locked (150 ms–467 ms) time window during which midfrontal theta power increased for nonpreferred relative to preferred outcomes. **p p S5 Text). Right: Contrast for nonpreferred versus preferred outcomes. The significant (nonsignificant) a priori midfrontal channels are again indicated by white (gray) discs (post hoc Bonferroni corrected alpha = 0.017). (b) Feedback-related theta power did not vary as a function of biased learning. Left: Midfrontal theta power within the resulting time window did not significantly increase for the conditions in which we observed biased learning at the behavioral level (rewarded Go: ε0 + κ and punished NoGo: ε0 − κ) relative to their unbiased counterparts (rewarded NoGo and punished Go: ε0; BF01 = 4.9). Right: Midfrontal theta power also did not significantly vary as a function of relative magnitude of instrumental learning. Here, we contrasted rewarded Go responses (enhanced learning: ε0 + κ) and punished Go with rewarded NoGo and punished NoGo (decreased learning: ε0 − κ; BF01 = 3.1). Underlying data can be found at http://hdl.handle.net/11633/di.dccn.DSC_3017033.03_624.